Development and Validation of a Machine Learning Algorithm for Predicting Diabetes Retinopathy in Patients With Type 2 Diabetes: Algorithm Development Study
Bibliographic record
Abstract
Background: Diabetic retinopathy (DR) is the leading cause of preventable blindness worldwide. Machine learning (ML) systems can enhance DR in community-based screening. However, predictive power models for usability and performance are still being determined. Objective: This study used data from 3 university hospitals in South Korea to conduct a simple and accurate assessment of ML-based risk prediction for the development of DR that can be universally applied to adults with type 2 diabetes mellitus (T2DM). Methods: DR was predicted using data from 2 independent electronic medical records: a discovery cohort (one hospital, n=14,694) and a validation cohort (2 hospitals, n=1856). The primary outcome was the presence of DR at 3 years. Different ML-based models were selected through hyperparameter tuning in the discovery cohort, and the area under the receiver operating characteristic (ROC) curve was analyzed in both cohorts. Results: Among 14,694 patients screened for inclusion, 348 (2.37%) were diagnosed with DR. For DR, the extreme gradient boosting (XGBoost) system had an accuracy of 75.13% (95% CI 74.10-76.17), a sensitivity of 71.00% (95% CI 66.83-75.17), and a specificity of 75.23% (95% CI 74.16-76.31) in the original dataset. Among the validation datasets, XGBoost had an accuracy of 65.14%, a sensitivity of 64.96%, and a specificity of 65.15%. The most common feature in the XGBoost model is dyslipidemia, followed by cancer, hypertension, chronic kidney disease, neuropathy, and cardiovascular disease. Conclusions: This approach shows the potential to enhance patient outcomes by enabling timely interventions in patients with T2DM, improving our understanding of contributing factors, and reducing DR-related complications. The proposed prediction model is expected to be both competitive and cost-effective, particularly for primary care settings in South Korea.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".